Retroviruses such as HIV are distinguished from other viruses by two steps in the viral life cycle, reverse transcription of the viral RNA to make a cDNA copy, and integration of that cDNA into a chromosome of the host. The reverse transcription step has been exploited as a target for antiviral agents. Another HIV encoded enzyme, the protease, has also been exploited as an inhibitor target. Inhibitors of these steps, however, have been found to be plagued by viral escape mutants in clinical applications, adding interest to the search for new inhibitors of other targets. We propose to develop inhibitors of the third HIV-encoded enzyme, integrase, a viral function that has not been exploited as an inhibitor target. We will use a combination of informed screening of new compounds, structure- based design, and iterative synthesis to created effective inhibitors. We have improved the X-ray structure of the catalytic domain of integrase to 1.8 Angstroms resolution, providing key background for structure-based design. We have in place a series of assays for inhibition ranging from reactions with purified proteins to growth of HIV in culture. Already we have used such assays to identify new inhibitors of integration. We propose to form cocrystals of integrase with inhibitors, then use the results obtained to synthesize further improved derivatives. Structures will then be obtained of integrase bound to the new compounds. In parallel assays will be carried out in vivo to tent efficacy and toxicity. By going through several cycles of cocrystallization new compound synthesis. and efficacy testing we hope to create useful small molecule inhibitors of HIV-1 integrase.